Virtual Machines as a Service

GPU-accelerated virtual machines and block storage for any AI and ML workloads. Provide secure and scalable computing capacity for hosting, testing and prototyping projects.

End User Self Service

Simple and Intuitive
End user self-service experience to configure, launch and use a VM with GPUs.

Latest GPU Models

From NVIDIA, AMD, and Intel
Support for state-of-the-art GPU models such as H100, A100, and more.

Scaling

Multiple GPUs per VM
Scale effortlessly from one to eight GPUs in a single virtual machine.

AI/ML Ready OS

Quick Deployments
Accelerate setup with pre-installed AI/ML images, including GPU and network drivers.

Block Storage

Integrated Storage
Turnkey integration with Ceph, DDN, Weka, Vast Data, Dell PowerStore, and others.

Integrated Metrics

Real-Time Usage Monitoring
View system performance and AI-specific metrics from the user console.

Metering

Accurate Resource Accounting
Precise, per-second metering for tracking VM resource usage to support billing and reporting.

VPCs

Private Network Support
Launch VMs inside isolated Virtual Private Clouds with unique CIDR blocks.

Public IPs

Direct Internet Connectivity
Assign public IPs to VMs for external access—no bastion or NAT required.

High Performance

Nvidia Reference Architecture Compliant
Implements and complies with Nvidia's RA for high performance VMs.

vGPU Support

GPU Virtualization for Shared Access
Enable GPU sharing across multiple virtual machines using NVIDIA vGPU etc.

SXM GPU Support

Maximum Bandwidth & Performance
Leverage NVIDIA SXM-based GPUs (e.g., H100 SXM, A100 SXM) for extreme performance and bandwidth.